Magnetic field gradient
Citation, DOI, disclosures and article data
At the time the article was created Zach Drew had no recorded disclosures.View Zach Drew's current disclosures
At the time the article was last revised Jeremy Jones had no recorded disclosures.View Jeremy Jones's current disclosures
Signal localization for image construction in MR is based on adding a magnetic field gradient onto the main (constant) magnetic field.
In 1973, Paul Lauterbur published the idea in Nature of deliberately superimposing linear field gradients on the main magnetic field. Along each gradient, the signals would have different frequencies and could be plotted using a Fourier transformation (FT).
By passing current through gradients created by coils of wire (gradient coils), the magnetic field strength is altered in a controlled and predictable way. Gradients add or subtract from the existing field in a linear fashion, so that the magnetic field strength at any point along the gradient is known. At the isocenter the field strength remains unchanged – a certain distance away from the isocenter the field strength either increases (positive) or decreases (negative).
The strength/amplitude of the gradient is determined by the amount of current applied to the gradient coil (maximum amplitude determines maximum achievable resolution). Polarity determines which end of the gradient is positive and which is negative. This can be altered by changing the direction of the current in the coil (clockwise/anti-clockwise). The speed at which the gradient can be turned on (rise time) and turned off (slew rate) – which in turn determines the maximum scan speed of the system.
Imposing a gradient magnetic field changes both the precessional (Larmor) frequency and precessional phase of magnetic moments in a linear fashion. Faster magnetic moments gain phases compared to their slower neighbors.
The above characteristics can be used to encode the MR signal in three dimensions, using three orthogonal sets of gradients within the bore of the magnet. Gradients exist in the z, y and x axes with the isocenter at the center of all three gradients.
To spatially encode the image, 3 separate functions are necessary, with each gradient performing one of the tasks:
- slice select: locates a slice in the scan plane selected
- frequency encoding: locate a signal along the long-axis of the image
- phase encoding: locate a signal along the short-axis of the image
- 1. Jerrold T. Bushberg, John M. Boone. The Essential Physics of Medical Imaging. (2012) ISBN: 9780781780575
- 2. Lauterbur P. Image Formation by Induced Local Interactions: Examples Employing Nuclear Magnetic Resonance. Nature. 1973;242(5394):190-1. doi:10.1038/242190a0